I recently had my ECU rebuilt after rough idle and p1368 p0352 p0353 p0355 p0358 codes. The car runs better now with no codes but will often not start unless I give part throttle. Seems to happen randomly, after long drives/short drives/short parking/sitting overnight. I checked the Schrader valve: no squirt when cold, squirts with ignition on. Fuel pressure gauge on order. Runs perfectly after it starts. Any ideas on what to look for? I also notice that the car shutters when coming to a stop now, like she wants to keep moving. While sitting at a light I will get waves of surges where the tone of the engine changes and I feel a little shutter. I think these are two separate issues, but thought I'd give as much info as I could. Thanks.

 

P1368 and other codes refer to fault with the ignition module bank 2 that fires plugs on cylinders 2, 3, 5 & 8.

Ignition monitoring circuit between splice and
ECM: open circuit, short circuit to ground,
short circuit to B+ voltage
Ignition module / coil bank 2 ground circuit
fault

Make sure your battery is in good shape and fully charged. You can switch the two ignition modules located up on the firewall, middle/rear of engine compartment. Clear codes and then see if the P0367 error crops up for the other 4 cylinders. You can hope disconnecting & reconnecting wire connectors improves the issue.

 

jrnsr - Thanks for your answer, but as I said in the post the ECU rebuild took care of the codes. This is not the answer.

 

I am thinking a stuck open inline fuel return check valve since you are losing residual line pressure (no squirt when cold). I am not an expert, maybe someone more knowledgable will respond.

 

Or a leaky injector.

 

Maybe I missed it, but what's the model and year?

Rough idle + misfires => new ECU + minor peculiarities (starting and inconsistent RPM issues)
That all might be explained by poor Throttle Position Sensor adjustment. Was the TPS ever adjusted prior to or after ECU replacement?
I've been playing with the TPS on our '99 XK8 and will be replacing it soon. I want to document some dimensions, readings and an adjustment procedure I've toyed with and will have more specific suggestions in the near future.

 

I've got these same 5 codes coming up on my 2000MY XK8. There are no ignition modules on my car as many suggest checking, the ignition modules are fitted on earlier cars with 2-pin coils but deleted when 4-pin coils were introduced. Each 4-pin coil has its own module built in. This same fault (P1367 or P1368) coming up tends to immediately suggest problem is one of the 2 ignition monitors/modules, however, the same 2 sets of 4 coils/cylinders are still grouped together as they were connected to ignition modules 1 or 2 on the earlier cars, this should explain the confusion. The same 2 sets of 4 coils feed back to 2 separate pins at the ECU.
So as I cant find any other reason why I have these faults, I'm suspicious of a problem inside the ECU. Can you tell me what they found wrong in yours, sounds like they fixed it though :-)

 

SHEJAGinNZ - Open the ECU and look for something like this, see the center of the picture and corroded leads. The fix is to replace the old capacitors and repair the traces. I can't recommend the company that fixed mine, there were screws missing upon return, but there are other repair shops out there. I won't share the name of my company because the ECU works and it would be unfair to slam the company for what might be a one time error. Good luck.

 

Matt-97 - Wow, awesome, thanks. Have looked at mine but didn't see anything untoward (see pics). But will open up again and take another look.

 

I can see nothing obviously adrift in your ECM pics.
The 'leaky capacitor' syndrome seems to be largely limited to the '97 ECM, for some obscure reason.

Whilst the ignition modules were deleted and their functionality integrated into the ignition coils at the changeover to the AJ27 engine, the operation is basically the same. The ECM monitors the 'spark OK' feedback from each ignition coil.

Here are the DTCs for the 5 codes* I've also included P0351 as the default action/possible causes refer to it:




The eight 'spark OK' feedback signals from the ignition coils are combined into groups of four, two from each bank per group. The P1368 code is flagging that the ECM can't see proper feedback from group two (1B, 2A, 3A, 4B):- the P035x codes are for the individual ignition coils in that group.

I would check the posssible causes listed in the P1368 row. You may have a problem in the engine loom - short/open in the feedback line, or it's possible that a faulty ignition coil could be dragging the whole group down. Perhaps try disconnecting each coil in the group to see if that makes the other codes go away.


* My take on 'ignition module' here is that it is the same as 'ignition coil' for the AJ27.

 

michaelh - thank you very much for your input, and great information, most of which I understand, but a couple of interesting points there I need to check.
I have swapped out each of those 4 coils one at a time with the new coil I have, but I haven't actually tried disconnecting one coil at a time as you suggest... will try that.
I have however confirmed that the 2 groups of 4 coils each (1B, 2A, 3A, 4B & 1A, 2B, 3B, 4A) feedback wiring is good back to the ECU, one group goes back to pin #10 at plug EM83, the other group goes to pin #11.

 

Oops! I'm wrong about the transfer of functionality from ignition module to ECM - strictly that should be to the ignition coils. I've amended my post.

It appears you've already checked most of the possible causes listed:- are the coil grounds all OK & no shorts to B+ or ground on the splice?

My interpretation of the code monitoring conditions is that the ECM allows a brief period after start-up to allow the motor to settle down before taking notice of the feedback. Does the engine initially run smoothly for a few seconds?

Your coil substitution tests appear to rule out an ignition coil failure, but I'd still do the disconnect one at a time just in case you're really unlucky and the new coil has an issue, Either that or check it in one of the functioning cylinders to confirm it's OK.

Any chance you have access to an oscilloscope?

 

Yes, I was going to say something about the ignition module functionality transfer you mention.

Ok, I need to check the coil grounds, splice grounding etc. Will do that next day or so, along with one by one coil disconnect procedure.

Yes, you are correct regarding smooth running initially after start-up, then begins rough running and 'Restricted performance' notification comes up.

And, look what I have in my garage... it was my dads. Don't ask me how it works though, although it would be good to learn.

 

I frequently prop my cell phone up on the console running Torque lite. I peek at the MPG, voltage, water temp and other paramateres I have questions with. One I visit frequently is throttle position to notice where it is at before startup, how it shuts down warming up or how much it compenstes with accessories on.. Usually it show 5+% cold and drops to 1-2% warmed up.

It could be the new ECU is interpreting the TPS readings a few percent differently than the old unit. I like the bluetooth OBD scanner over the wired one, since I can place it where it is visble in the engine compartment when fiddling.

Why not check to see what the scanner reads with the throttle closed completely. I'll bet it shows a few degrees open with the butterfly shut tight. The ECM reads what it thinks is 5-7% open when it it really around 2 or 3%, so you have to compensate with your foot to feed more air. Or maybe it is short changing the fuel thinking it it is closed more than it really is. Google "TPS and PPS adjustment via ODB" for details to tweak Throttle Position Sensor, Pedal Position Sensor and ECM to work together.

I also like to use my cell phone as an oscilloscope, but that's another story.

What the heck are "Spoilers" (below)? Better yet, how does one get rid of them?


 

We've discussed the newer coils with the built in modules and feeedback to the ECM. Now if you ask a mechanic how it works, he'd tell you to call the dealership. Dealership would say you need to talk to Jaguar in the UK for details. Jaguar would feign ignorance and tell you it was designed and built by Ford, so call Dearborn Michigan Ford electrical engineers. So I talked to one of the electrical engineers I'd crossed paths with in our MG club and he said they used that particular coil with only 4 wires, so he didn't know what was going on with the feedback circuit. He'd talk to the guys in the department that designed and built them. He later replied that most of those engineers are long gone- Ford was downsizing their department shifting focus to electric cars. I took some photos of my cell phone oscilloscope displaying the waveforms that revealed the handshaking between each coil and the ECM and emailed them to him. That was last year; a lot of time has passed and I don't have any recollection of what we discussed!

 

A drop in brand spanking new replacement ECU for our cars would be really nice. Swallows racing do a an Emerald K6 swap, probably not cheap though. Fixing our own ECU's is the cheapest route these days still though. It would also be nice to see a manual swap kit complete, I know Swallows will also do that for you but once again I doubt it's "cheap".

 

Moral of the story is don't post well after midnight and a couple of glasses of red (just back from Madeira & still in holiday mode).

It would be better if this was moved to a new thread, as we've strayed OT from the original title and the problems of leaky caps in 97 ECMs.

If you can coax some life out of the old Kikusui then we should be able to see what the ECM is seeing. Try to get a green horizontal line to display, with the 'sweep range' on the 1 or 10 KHz setting and juggling the 'position' controls.

As one cylinder group is behaving, there's a reference waveform to compare with.

 

I'll give 'the old Kikusui' (that I still remember my dad buying many years ago) a cleanup later today, and see how it looks